BPA in Premature Infants

Study Examining BPA Exposure in Premature Infants by Calfat et al.

A study by Calfat et al. published in the January 2009 edition of Environmental Health Perspectives describes a cross-sectional study in which urine samples from premature infants in two Boston-area hospitals were analyzed for free and total BPA and other compounds.1 This critique will focus only on the results pertaining to BPA. A total of 41 infants had at least one urine sample available for analysis of BPA and the authors were able to analyze both free and total BPA in 37 infants. A key finding from this study is that the geometric mean of the urinary total BPA for this sample (30.3 mcg/L) was an order of magnitude higher than the urinary total BPA levels observed in 6-11 year-old children from NHANES 2003-2004 (median = 3.7 mcg/L). However, the source of the increased BPA exposure in the premature infants could not be identified from the study, although a source could be medical tubing and IV bags used to treat premature infants. The authors of the study compared the two hospitals and found that the total urinary BPA levels in infants from one hospital were significantly higher than from the other hospital, suggesting wide variability in exposure of premature infants.

Another key finding is that the geometric mean for free, non-glucuronidated BPA was 1.8 mcg/L, suggesting that approximately 95% of the BPA in the pre-term infants was glucuronidated. It is well established that free BPA can potentially interact with estrogen receptors, but glucuronidation of BPA represents a detoxification pathway. This study sheds new light on the previously thought notion that premature infants and term neonates possess greatly reduced capacity for BPA glucuronidation relative to adults. The authors referred to this 95% BPA glucuroidation rate as evidence that preterm infants possess "some capacity to metabolize BPA," although this seems to be an understatement in light of the data presented.

In this study, it is possible that the "true" level of free bisphenol A in the urine was even lower than the reported values for two reasons. First, the authors did not prescreen their sample collection systems or laboratory/analytical equipment for BPA contamination. BPA is a widespread environmental contaminant and can be found in water, house dust, and other common materials. Second, it is conceivable that at least a small percentage of conjugated BPA in the urine samples would undergo deconjugation during sample storage. The authors cited their unpublished data that the conjugated BPA in urine samples was stable for up to 30 months when stored at subfreezing temperatures, but no data were presented to indicate the criteria used to define "stable." Thus, it is not known what percentage of loss was considered acceptable. Given the trace levels of this compound in the urine and the claims of some individuals that BPA in even parts per billion (ppb) or parts per trillion (ppt) concentrations may have adverse health effects, extra care is required regarded the reporting of the sample storage and analytical methods used in BPA studies.

In addition to the above finding, there was a strong linear correlation between free and total BPA (Spearman r = 0.86). Thus, the authors found that the degree of BPA glucuronidation increased as did BPA exposure. Of the 37 infants having measures of both total and free BPA in the urine, 32 of them had levels of free BPA that were 10 mcg/L or less. The 5 infants with free BPA levels greater than 10 mcg/L (the highest was 17.3 mcg/L) also had the highest total BPA exposure (the highest was 946 mcg/L). It is not clear what was responsible for the highest levels of BPA exposure that were observed, but it is clear that BPA glucuronidation was greatly increased in these infants.

These data support the recent assessment by EFSA that infants can glucuronidate BPA doses up to 1 mg/kg bw/day2 and contrast the results of a previous study in neonatal mice reporting that the method of BPA administration (subcutaneous versus oral) was of no significance in toxicological studies with infants, presumably because of the infant's lack of ability to glucuronidate BPA.3 The existence of data in human premature infants showing near total glucuronidation of BPA, at even relatively high exposures, is of greater scientific value than pharmacokinetic modeling studies based on assumptions from data that present an incomplete picture of UDP-glucuronsyltransferase (UGT) enzyme(s) responsible for BPA glucuronidation,4 or rodent studies such for assessing the toxicity potential of BPA.

There are several limitations of this study. The study was cross-sectional, relying on spot urine samples versus 24-hour urine collections. The half-life of BPA in the body is relatively short (~2-5 hours in adults) and, thus, recent exposures to BPA relative to the collection of the urine samples could bias the results toward greater exposure levels. A second limitation is that serum or plasma BPA measurements were not obtained, which, though understandable given the vulnerability of the population studied, nonetheless limits the usefulness of the data. Blood levels of BPA would be valuable in assessing exposure of the various tissues in the body to free BPA. A third limitation is that there are no follow-up data to report health outcomes of the infants that were exposed to varying levels of BPA and the study was not designed to address that question. This study provides at least some indication that the human premature infant is better at detoxifying BPA than previously assumed.

###

References Cited:

1. Calfat AM, Weuve J, Ye X, Jia LT, Hu H, Ringer SR, Huttner K, Hauser R. Exposure to bisphenol A and other phenols in neonatal intensive care unit premature infants.
Environ Health Perspect 2008. doi:10.1289/ehp.0800265 (available at http://dx.doi.org).

2. European Food Safety Authority. 2008. Toxicokinetics of Bisphenol A: Scientific Opinion of the Panel on Food Additives, Flavourings, Processing Aids, and Materials in Contact with Food (AFC). Question no. EFSA-Q-2008-382.
EFSA Journal
759:1-10.

3. Taylor JA, Welshons WV, vom Saal FS. 2008. No effect of route of exposure (oral;subcutaneous injection) on plasma bisphenol A throughout 24 hr after administration in neonatal female mice.
Reprod Toxicol
doi: 10.1016/j.reprotox.2008.01.001.

4. Edginton AN, Ritter L. Predicting plasma concentrations of bisphenol A in young children (<2 years) following typical feeding schedules using a physiologically-based toxicokinetic model.
Environ Health Persp
2008. doi: 10.1289/ehp.0800073 (available at http://dx.doi.org)